I’ve created a bot which uses the API to reply to questions. The answers are basically a selection of potted answers depending on the state of the interaction with the other party.
For the first day, it was working fine, then overnight replies began to be hidden. If I navigate to the timeline of my bot, the replies are definitely there, and the original human-originated tweet shows there is one reply!
But the replies don’t always show up, and the originator doesn’t get a notification that the bot has replied. Occasionally (very rarely) one does, but 90% of the time, the tweet exists but is invisible unless you visit the bot’s Tweets & Replies page…
The only rule I can see that’s being bent is this one, which may be applied incorrectly in this case:
- if you post duplicative or substantially similar content, replies, or mentions over multiple accounts or multiple duplicate updates on one account, or create duplicate or substantially similar accounts;
Please do not:
- Repeatedly post duplicate or near-duplicate content (links or Tweets).
- Send automated Tweets or replies.
- Use bots or applications to post similar messages based on keywords.
- Post similar messages over multiple accounts.
So, what’s the point in having an API that can read and reply to messages automatically, if automatically reading and responding is not permitted?!
Yes, if people ask the same question several times, the response will be the same… But we’re talking about two or three times per person who interacts with the account…
How do I actually use the API to make a responsive bot?!?